Validity of self-reported diabetes varies with sociodemographic charecteristics: Example from Iran

Validation of self-reported medical condition in the Taiwan Biobank

BACKGROUND

This study aimed to validate self-reported medical conditions in the Taiwan Biobank (TWBB), in which participants were inquired about 30 disease conditions, by comparing them with claims records from Taiwan's National Health Insurance (NHI) claims database.

METHODS

We identified 30 clinical diagnoses using ICD-CM codes from ambulatory and hospital claims within the NHI claims database, matching diseases included in the TWBB. The concordance between self-reports and claims records was evaluated using tetrachoric correlation to assess the correlation between binary variables.

RESULTS

A total of 131,834 participants aged 30-70 years with data from the TWBB and NHI records were included. Concordance analysis revealed tetrachoric correlations ranged from 0.420 (chronic obstructive pulmonary disease) to 0.970 (multiple sclerosis). However, several disorders exhibited lower tetrachoric correlations. The concordance was higher among those with higher education attainment, and lower among married individuals.

CONCLUSION

The concordance between self-reports in the TWBB and NHI claims records varied across clinical diagnoses, showing inconsistencies depending on participant characteristics. These findings underscore the need for further investigation, especially when these variables are crucial to research objectives. Integrating complementary databases such as clinical diagnoses, prescription records, and medical procedures can enhance accuracy through customized algorithms based on disease categories and participant characteristics and optimize sensitivity or positive predictive values to align with specific research objectives.

An Examination of the Agreement between Self-reported Diabetes and Paraclinical Tests, Medical Records, and Clinical Examinations: Insights from the Shahedieh Cohort Study, Yazd, Iran

Introduction

Many researchers utilize self-reports to evaluate the prevalence of diseases. However, the accuracy of these self-reports remains uncertain in various studies. The objective of this particular study was to validate self-reported cases of diabetes among adults aged 35-70 years participating in the Shahedieh Cohort Study (SHCS).

Methods

This cross-sectional study was conducted using data from the first phase of SHCS during 2015-2017. The study included 1000 Iranian adults aged 35-70 years. The Gold standard for diabetes was determined by measuring fasting blood sugar (FBS) and evaluating the history of treatment and use of diabetes medications. To assess self-report validity, various statistical indices such as sensitivity, specificity, positive and negative predictive values (NPVs), accuracy, positive likelihood ratio (LR+), negative likelihood ratio (LR-) and AUC were used. The agreement between self-reported diabetes and the gold standard was assessed using kappa statistics. All statistical analyses were performed using SPSS version 13 and R 4.3.1 software.

Results

The study findings indicated that the prevalence of diabetes was 18% according to self-report and 19.9% according to the gold standard measurement. The self-report accuracy was 95.67%. There was perfect agreement (kappa=0.86) between the self-report and gold standard criteria. The AUC, sensitivity and specificity of self-reported diabetes were 0.937, 93.82% and 96.08%, respectively. Additionally, the results suggested that the self-report of diabetes was more valid in individuals with a normal body mass index (BMI) and without a family history of diabetes in first-degree relatives.

Conclusion

The results showed that in the absence of diabetes control programs, self-report of diabetes is reliable and recommended.

Prevalence of Hypoglycemia and Its Determinants Among Diabetes Patients on Insulin Treatment at Tepi General Hospital, Southwest, Ethiopia

Background

Hypoglycemia occurs when the blood sugar levels are too low. In severe cases, hypoglycemia may give to seizures, coma, and sometimes death. The prevalence of hypoglycemia among patients with diabetes is likely underreported and undocumented.

Methods

An institutional-based cross-sectional study was conducted among 336 study participants at the Tepi General Hospital in Southwest Ethiopia. A binary logistic regression model was used to determine the association between the prevalence of hypoglycemia and the factors associated with it. After a bivariate analysis, variables with a p value of < 0.25 were selected as a candidate for multivariable analysis. An odds ratio with a 95% CI was considered to indicate a significant association.

Results

With 95% CI (52.48-62.25) the prevalence of hypoglycemia was determined as 57.44%. Nearly 11% of the patients encounter severe hypoglycemia. Female patients had three times higher odds of experiencing hypoglycemia (aOR=3, 95% CI: 2.10, 6.39). Those with no formal education had 1.5 times higher odds of experiencing hypoglycemia (aOR = 1.5, 95% CI: 1.08, 5.45). Patients with type 1 diabetes were 3.4 times more likely to experience hypoglycemia (aOR = 3.4, 95% CI: 1.87, 7.50). Those who had been diagnosed before 10 years (aOR = 1.3, 95% CI: 1.02, 3.21) were more likely to have hypoglycemia. Furthermore, patients who consumed alcohol were 3.0 times more likely to have a history of hypoglycemia (aOR = 3.0, 95% CI: 2.03, 6.43).

Conclusion

The magnitude of hypoglycemia among patients with diabetes is determined to be considerable with more than half of the participants encountered hypoglycemia. There is a strong relationship between the occurrence of hypoglycemia and sex, type of diabetes, time since diagnosis, alcohol status, and education status. Therefore, all concerned parties must pay close attention to lessen the prevalence of hypoglycemia and address the problems based on the primary contributing factors.

Validation of the self-reported diagnosis of diabetes mellitus, hypercholesterolemia, and hypertension in Iran; STEPS 2016

Purpose

As a part of STEPwise approach to risk factor Surveillance (STEPS) study, our aim was to evaluate the validity of the self-reported diagnosis of diabetes (DM), hypertension (HTN), and hypercholesterolemia (Hyper-Chol) in the Iranian population.

Methods

Using systematic proportional to size cluster sampling, 27,232 participants were included in our study. We calculated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) to assess the validity of self-reported diagnoses. Furthermore, logistic regression was employed to examine the relationship between the validity of self-reported diagnoses and sociodemographic and lifestyle factors. All analyses were performed using STATA version 14.

Results

The PPV for self-report of DM, HTN, and Hyper-Chol were estimated to be 69%, 74% and 80%, and NPV measured up to 95%, 84%, and 50%, respectively. Positive/negative self-reports were more accurate among older (younger) individuals. Age had a negative correlation with the validity of self-reported Hyper-Chol but a positive correlation with the validity of self-reported DM and hypertension HTN. Additionally, an increase in BMI was associated with an increase/decrease in PPV and a decrease/increase in NPV across all diseases.

Conclusion

Self-report studies hold value in situations where direct in-person interaction is not feasible, either due to prohibitive costs or restrictions imposed by infectious diseases (COVID-19). Self-report surveys are valuable tools in studying the epidemiology of diseases; however, the type of the disease, the study purpose, either finding sick people or healthy people, the age subgroups, and socioeconomic status should be taken into consideration.

Identifying factors associated with of blood pressure using Structural Equation Modeling: evidence from a large Kurdish cohort study in Iran

BACKGROUND

Identifying the risk factors leading to hypertension can help explain why some populations are at a greater risk for developing hypertension than others. The present study seeks to identify the association between the risk factors of hypertension in 35- to 65-year-old participants in western Iran.

METHODS

This cross-sectional study was conducted on 9705 adults from baseline data of Ravansar Non-Communicable Disease (RaNCD) cohort study, in the west region of Iran. Each of the latent variables were confirmed by confirmatory factor analysis. Using Structural Equation Modeling (SEM), we assessed the direct and indirect effects of factors associated with blood pressure.

RESULTS

Socioeconomic status (SES), physical activity, mean of serum lipids, obesity, diabetes and family history of hypertension had a diverse impact on the blood pressure, directly and (or) indirectly. The standardized total effect of SES, physical activity, mean of serum lipids, and obesity were -0.09 vs. -0.14, -0.04 vs. -0.04, 0.13 vs. 0.13 and 0.24 vs. 0.15 in men and women, respectively. Diabetes had a direct relationship with the blood pressure in women (0.03).

CONCLUSION

With regard to control of high blood pressure, public health interventions must target obesity, lifestyle and other risk related to nutritional status such as hyperlipidemia and hyperglycemia in Iranian population and among those with higher SES.

Socio-demographic correlates of diabetes self-reporting validity: a study on the adult Kurdish population

BACKGROUND

In this research, data of the DehPCS study were used to assess the validity of self-reported diabetes based on the reference criteria, including the history of taking oral anti-diabetic drugs, insulin injection, or high fasting blood sugar.

METHODS

A cross-sectional analytical study was performed on 4400 participants of the DehPCS study, aged 35-70 years. The reference criteria were oral hypoglycemic drug consumption, insulin injection, and/ or fasting blood sugar ≥126 (mg/dl). The self-reporting diabetes was investigated by well-trained interviewers before the diabetes diagnosis based on the reference criteria. The validity of self-reporting diabetes was assessed using sensitivity, specificity, as well as positive and negative predictive values. Socio-demographic correlates of self-reported agreement were examined by multinomial logistic regression.

RESULTS

Three thousand nine hundred ninety-six people participated in this study, and the participation rate was equal to 90.8%. The diabetes prevalence among the study population was 13.1% based on self-reports and 9.7% based on the reference criteria. Five hundred twenty-three participants reported diabetes, 213 (41.28%) of whom did not have it. We found a good agreement of 92.3% with an acceptable kappa value of 65.1% between self-reporting diabetes and the reference criteria. Diabetes self-reporting also guaranteed sensitivity of 78.5%, specificity of 93.9%, as well as the positive and negative predictive values of 58.7% and of 98.0%, respectively. Being female, the higher economic class, the higher body mass index (BMI), and the positive family history of diabetes increased the chance of false positive. Being male, older ages and the moderate economic class increased the chance of false positive.

CONCLUSION

Self-reporting diabetes is identified as a relatively valid tool which could fairly determine the diabetes prevalence in epidemiological studies. It should be noted that its validity is influenced by some socio-demographic characteristics.

Self-Reported Modifiable Risk Factors of Cardiovascular Disease among Older Adults in Malaysia: A Cross-Sectional Study of Prevalence and Clustering

The co-occurrence of multiple modifiable risk factors increases the risk of cardiovascular disease (CVD) morbidity or mortality. This study examines the prevalence and clustering of self-reported modifiable CVD risk factors among older adults in Malaysia. A total of 7117 adults aged ≥50 years participated in the National Health and Morbidity Survey (NHMS) 2018: Elderly Health, a community-based cross-sectional survey. Data were collected using a standardized structured questionnaire. Multivariable logistic regression was used to determine the factors associated with the clustering of self-reported modifiable CVD risk factors. The prevalence of self-reported diabetes, hypertension, hypercholesterolemia, overweight/obesity, and current smoking was 23.3%, 42.2%, 35.6%, 58.4%, and 17.5%, respectively. Overall, the prevalence of clustering of ≥1, ≥2, and ≥3 modifiable CVD risk factors was 83.3%, 75.4%, and 62.6%, respectively. Multivariable logistic regression analysis showed that men, 60–69 age group, urban dwellers, having no formal education, unemployed/retirees/homemakers, and being physically inactive were independently associated with self-reported modifiable CVD risk factors clustering. There are also ethnic differences in self-reported modifiable CVD risk factors clustering. Our findings underscore the necessity of targeted interventions and integrated strategies for early detection and treatment of modifiable CVD risk factors among older adults, considering age, sex, ethnicity, and socioeconomic status.

Self-reported diabetes, hypertension and hypercholesterolemia among older persons in Malaysia

AIM

The objective of this study is to determine the prevalence of self-reported diabetes, hypertension and hypercholesterolemia among older persons in Malaysia and their associated factors.

METHODS

The study used data from the National Health and Morbidity Survey in 2018. It was a cross-sectional study with two-staged stratified cluster sampling design. In total, 3977 adults aged ≥60 years were selected for this study. Respondents were interviewed face to face using a structured questionnaire. Self-reported diabetes, hypertension or hypercholesterolemia was defined as having ever been told they have these diseases by a medical doctor or paramedic. Data were analyzed using SPSS version 25. The multiple logistic regression model was used to examine the factors associated with the prevalence of self-reporting.

RESULTS

The prevalence of self-reported diabetes, hypertension and hypercholesterolemia among older persons in Malaysia were 27.7%, 51.1% and 41.8% respectively. Presence of other comorbidities and being obese showed higher odds for all three diseases. Indians, unemployed, inactive had higher odds for diabetes. Other Bumiputras, unemployed, non-smoker, obese and inactive had higher odds for hypertension. Non-smoker had higher odds for hypercholesterolemia.

CONCLUSIONS

Health promotion, vigilance, attention and services targeting on the associated factors should be strengthened for older persons in Malaysia to ensure healthy aging. Geriatr Gerontol Int 2020; 20: 79-84.